The impact on global warming potential of converting from disposable to reusable sharps containers in a large US hospital geographically distant from manufacturing and processing facilities Brett McPherson 1 2 3
1
, Mihray Sharip
2
, Terry Grimmond Corresp.
3
Director, Environmental Health, Loma Linda University Health, San Bernardino, CA, United States Environmental Health Specialist, Loma Linda University Health, San Bernardino, CA, United States Director, Grimmond and Associates, Microbiology Consultancy, Hamilton, New Zealand
Corresponding Author: Terry Grimmond Email address:
[email protected]
Background. Sustainable purchasing can reduce greenhouse gas (GHG) emissions at healthcare facilities (HCF). A previous study found that converting from disposable to reusable sharps containers (DSC, RSC) reduced sharps waste stream GHG by 84% but, in finding transport distances impacted significantly on GHG outcomes, recommended further studies where transport distances are large. This case-study examines the impact on GHG of nation-wide transport distances when a large US health system converted from DSC to RSC. Methods. The study examined the facility-wide use of DSC and RSC at Loma Linda University Health (LLUH), an 1100-bed US hospital system where: the source of polymer was distant from the RSC manufacturing plant; both manufacturing plants were over 3,000 km from the HCF; and the RSC processing plant was considerably further from the HCF than was the DSC disposal plant. Using a “cradle to grave” life cycle assessment (LCA) tool we calculated annual GHG emissions (CO2, CH4, N2O) in metric tonnes of carbon dioxide equivalents (MTCO2eq) to assess the impact on global warming potential (GWP) of each container system. Primary energy input data was used wherever possible and region-specific impact conversions used to calculate GWP of each activity over a 12-month period. Unit process GHG were collated into Manufacture, Transport, Washing, and Treatment & disposal. Emission totals were workload-normalized and analysed using CHI2 test with P ≤0.05 and rate ratios at 95% CL. Results. Using RSC, LLUH reduced its annual GWP by 162.4 MTCO2eq (-65.3%; p < 0.001; RR 2.27-3.71), and annually eliminated 50.2 tonnes of plastic DSC and 8.1 tonnes of cardboard from the sharps waste stream. Of the plastic eliminated, 31.8 tonnes were diverted from landfill and 18.4 from incineration. Discussion. Unlike GHG reduction strategies dependent on changes in staff behaviour (waste segregation, recycling, turning off lights, car-pooling, etc), purchasing strategies can enable immediate, sustainable and institution-wide GHG reductions to be achieved. Medical waste containers contribute significantly to the supply chain carbon footprint and, although non-sharp medical waste volumes have decreased significantly with avid segregation, sharps wastes have increased, and can account for up to half of total medical waste mass. Thus, converting from DSC to RSC can assist reduce the GWP footprint of the medical waste stream. This study confirmed that large transport distances between polymer manufacturer and container manufacturer; container manufacturer and user; and/or between user and processing facilities, can significantly impact the GWP of sharps containment systems. However, even with large transport distances, we found that a large university health system significantly reduced the PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
GWP of their sharps waste stream by converting from DSC to RSC.
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
1
The impact on global warming potential of converting from disposable to reusable sharps
2
containers in a large US hospital geographically distant from manufacturing and
3
processing facilities
4 5
Brett McPherson1 BSN, Mihray Sharip2 MS, REHS, CHMM, Terry Grimmond3 FASM,
6
BAgrSc, GrDpAdEd&Tr
7
1Director
8
2Environmental
9
3Director,
Environmental Health, Loma Linda University Health, Loma Linda, CA, USA Health Specialist, Loma Linda University Health, Loma Linda, CA, USA
Grimmond and Associates, Microbiology Consultants, Hamilton, New Zealand
10 11
Corresponding author: Terry Grimmond, 930 River Rd Queenwood, Hamilton New Zealand 3210.
12
Em:
[email protected] Ph: +64 274 365 140
13 14
ABSTRACT
15
Background. Sustainable purchasing can reduce greenhouse gas (GHG) emissions at healthcare
16
facilities (HCF). A previous study found that converting from disposable to reusable sharps
17
containers (DSC, RSC) reduced sharps waste stream GHG by 84% but, in finding transport
18
distances impacted significantly on GHG outcomes, recommended further studies where
19
transport distances are large. This case-study examines the impact on GHG of nation-wide
20
transport distances when a large US health system converted from DSC to RSC.
21
Methods. The study examined the facility-wide use of DSC and RSC at Loma Linda University
22
Health (LLUH), an 1100-bed US 5-hospital system where: the source of polymer was distant
23
from the RSC manufacturing plant; both manufacturing plants were over 3,000 km from the
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
24
HCF; and the RSC processing plant was considerably further from the HCF than was the DSC
25
disposal plant. Using a “cradle to grave” life cycle assessment (LCA) tool we calculated annual
26
GHG emissions (CO2, CH4, N2O) in metric tonnes of carbon dioxide equivalents (MTCO2eq) to
27
assess the impact on global warming potential (GWP) of each container system. Primary energy
28
input data was used wherever possible and region-specific impact conversions used to calculate
29
GWP of each activity over a 12-month period. Unit process GHG were collated into
30
Manufacture, Transport, Washing, and Treatment & disposal. Emission totals were workload-
31
normalized and analysed using CHI2 test with P ≤0.05 and rate ratios at 95% CL.
32
Results. Using RSC, LLUH reduced its annual GWP by 162.4 MTCO2eq (-65.3%; p < 0.001;
33
RR 2.27-3.71), and annually eliminated 50.2 tonnes of plastic DSC and 8.1 tonnes of cardboard
34
from the sharps waste stream. Of the plastic eliminated, 31.8 tonnes were diverted from landfill
35
and 18.4 from incineration.
36
Discussion. Unlike GHG reduction strategies dependent on changes in staff behaviour (waste
37
segregation, recycling, turning off lights, car-pooling, etc), purchasing strategies can enable
38
immediate, sustainable and institution-wide GHG reductions to be achieved. Medical waste
39
containers contribute significantly to the supply chain carbon footprint and, although non-sharp
40
medical waste volumes have decreased significantly with avid segregation, sharps wastes have
41
increased, and can account for up to half of total medical waste mass. Thus, converting from
42
DSC to RSC can assist reduce the GWP footprint of the medical waste stream. This study
43
confirmed that large transport distances between polymer manufacturer and container
44
manufacturer; container manufacturer and user; and/or between user and processing facilities,
45
can significantly impact the GWP of sharps containment systems. However, even with large
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
46
transport distances, we found that a large university health system significantly reduced the GWP
47
of their sharps waste stream by converting from DSC to RSC.
48 49
INTRODUCTION
50
Healthcare activities account for 5.4% of greenhouse gas (GHG) emissions in the U.K. (NHS,
51
2016; DBEIS, 2017) and 7.6% in U.S. (Chung and Meltzer, 2009) and, in hospitals, more than
52
half of GHG emissions are derived from supply chain goods and services (Chung and Meltzer,
53
2009; NHS, 2016). Many hospitals are adopting green purchasing strategies to reduce their GHG
54
(Chung & Meltzer, 2009; NHS, 2017) – a position supported by the Alliance of Nurses for
55
Health Environments (ANHE, 2017). Replacing disposable products with reusables is such an
56
example (WHO-HCWH 2009, Unger et al., 2016; Karrlson and Ohman, 2005) and, as clinical
57
waste containers are in the top 20 contributors to the supply chain carbon footprint (NHS, 2017),
58
replacing disposable sharps containers (DSC) with reusable sharps containers (RSC) is
59
recommended (PGH, 2013). One life cycle assessment (LCA) found that converting from DSC to
60
RSC achieved a significant reduction in GHG however the hospital was close to where both
61
containers were manufactured and the authors’ sensitivity analysis found that transport distances
62
could significantly affect results and recommended that scenarios with large transport distances
63
be investigated (Grimmond and Reiner, 2012). Our case-study compares the annual impact on
64
the global warming potential (GWP) of converting from DSC to RSC at a large U.S. teaching
65
hospital system sited at nation-wide distances from manufacturing plants.
66 67
MATERIALS AND METHODS
68
Study Overview
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
69
Using established principles for assessment of the life cycle GHG emissions of goods and
70
services (British Standards Institute, 2011) we utilised a cradle-to-grave life cycle inventory
71
(LCI) and a product-system LCA tool developed specifically for sharps containers and
72
containing some 750 data cells (Grimmond and Reiner, 2012). In a before-after intervention
73
study, we compared the annual GWP of DSC and RSC usage at Loma Linda University Health
74
(LLUH), an 1100 bed university healthcare system with 5 general acute care hospitals and an
75
expansive outpatient clinic system, in Loma Linda, California. Review Board approval by LLUH
76
was waived as no patients, patient data or patient specimens were involved.
77
The LCI itemised all energy-using processes required by each containment system life-cycle as
78
implemented at LLUH. Scope 1, 2 and 3 processes were included in both study years. Unit process
79
GHG were collated into the following life-cycle stages: manufacture (of polymer and containers);
80
transport; washing (RSC); and treatment & disposal. The LCA assessed the GWP of all energy
81
used in these processes (vehicle fuel, gas, electricity, water supply and treatment) and in the
82
manufacture and life cycle of ancillary products (pallets, transport cabinets, cardboard boxes,
83
wash products). The boundary of the system studied, together with inputs, outputs and
84
exclusions, are shown in figure 1.
85
Data Sources
86
The following data sources were used in calculating GHG: DSC and RSC resin manufacture
87
(American Chemistry Council, 2010); primary energy input data for DSC and RSC container
88
manufacture (Clarion 2011) and RSC washing (Daniels, 2017); industry-specific data for DSC
89
autoclaving (Daniels, 2012); RSC and DSC transport (DEFRA, 2015); eGRID values for
90
California, Michigan and Illinois power generation (USEPA eGRID 2016); National data for
91
energy inputs for US water supply and treatment (Chini and Stillwell, 2018); Industry data for
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
92
manufacture of wash products (Nielsen et al, 2013; Shahmohammadi et al, 2017); Industry data
93
for manufacture of cardboard (NCASI 2017), representative data for manufacture of transporters
94
(USDOE, 2010); Industry-specific data for pallet LCA (DEFRA, 2010): and US national values
95
for incineration of DSC (USEPA WARM, 2018. The same database and values were applied to
96
the relevant unit processes in DSC and RSC systems. Emissions for RSC manufacturing were
97
calculated using a worst-case scenario based on the actual age of the manufacturer’s oldest, most
98
frequently used RSC still in service. Although it is theoretically possible for RSC to be
99
recertified for a further period when they reach their certified reuse expiration, for this study their
100
“end-of-life” was taken to be the number of years under the above worst-case scenario. The
101
GHG associated with manufacture of ancillary reusables (transport cabinets and pallets) were
102
calculated on a per trip basis using their expected life span. Data on container size, model
103
number, number used, and total Adjusted Patient Days (APD) (workload indicator) were
104
obtained from LLUH. Total polymer required for manufacture of DSC and RSC was determined
105
by weighing an example of each model of container and multiplying by the number of
106
containers. The conversion-transition period (2 years) was excluded to avoid data overlap.
107
Emission totals for each system’s annual use were workload-normalized using APD for the two
108
study years. Results were analyzed using WinPepi v11.65 (WinPepi, 2016). A Yates-corrected
109
χ2 test was used for the analysis of proportions. Statistical significance was set at P ≤ .05 and
110
rate ratios calculated using 95% confidence intervals.
111
System Function, Boundary, Allocation and Classification
112
The system function provided by the alternative products (DSC, RSC) was the supply of sharps
113
containers for the disposal of sharps waste (biological, chemotherapeutic, pharmaceutical) within
114
LLUH. The functional unit was the supply of each system for a one-year period. Sharps waste is a
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
115
sub-category of medical waste and comprises items capable of penetrating human skin (e.g. needles,
116
scalpels) which may have the potential to transmit infectious disease or pose a physical or chemical
117
hazard. Because of these hazards, at disposal, all sharps must be safely contained in either DSC or
118
RSC and transported to a treatment facility. With DSC the container is used once and the intact
119
container and contents are subjected to treatment (commonly autoclaving or incineration) prior to
120
landfill. With RSC, the container is automatedly decanted of its contents (which are treated and
121
disposed), and the reusable container is robotically cleaned and decontaminated, and reused a defined
122
number of times. The boundary of the system studied (Figure 1) included the energy required for the
123
following unit processes: raw material extraction; polymer manufacture and transport; container
124
manufacture and transport; transport of full containers to treatment facility; RSC processing-energy
125
(including water supply, water treatment, and wash products); treatment of DSC; transport of treated
126
DSC to landfill; and energy required for electricity generation and supply. Transport fuel processes
127
were calculated from well to wheel. Excluded from the system boundary were treatment of contents
128
(identical in both DSC and RSC), infrastructure and assets, and any inputs and outputs that
129
comprised less than 1% of mass or energy (British Standards Institute, 2011), or were not relevant
130
to GWP.
131
Allocations for emissions were based on a mass basis for polymer production, container production,
132
cardboard cartons, RSC wash products, DSC treatments, and RSC parts recycling. Other allocations
133
for emissions were as follows: transport (truck size and utilization by tonne.km); electricity
134
generation (kWh); water supply and treatment (litre); RSC processing energy (container); RSC
135
transporters (trip); and pallets (trip).
136
Global warming potential was the impact assessment category to which all inventory data was
137
allocated as it is well-known, commonly used and understood by healthcare facilities. A table listing
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
138
the raw data for all unit processes in the LCI, including flow, units, emission factors, total GWP, data
139
sources and data-representativeness, accompanies this publication.
140 141
RESULTS
142
DSC were manufactured in Crystal Lakes IL from US-sourced polypropylene polymer, nested in
143
cardboard containers, transported 3,200km to the hospital on wooden pallets, and autoclaved and
144
landfilled without shredding at Vernon CA, 130km from the hospital. The RSC were manufactured
145
in Greenville MI from polymer sourced in Korea, transported 3,500km in reusable, proprietary
146
transporter cabinets to LLUH, and decanted and processed at Fresno CA, 440km from the hospital.
147
A summary of results is presented in the Table.
148
To service LLUH in the baseline year, 48,460 DSC were manufactured from 50.6 tonnes of polymer
149
and required 8.2 tonnes of corrugated cardboard packaging for transport (see Table). The DSC used
150
did not contain recycled polymer. In California, biological sharps are commonly treated by non-
151
incineration technologies (e.g. autoclave) then landfilled; chemotherapeutic and pharmaceutical
152
sharps must be incinerated (and ash landfilled) – this requires transport interstate as there are no
153
licensed incinerators for such wastes in California. With DSC, this resulted in 31.8 tonnes of plastic
154
DSC being landfilled and 18.8 tonnes of DSC being incinerated (Table 1).
155
In the RSC year, 2,779 RSC were manufactured from 9.6 tonnes of acrylonitrile butadiene styrene
156
(ABS) polymer, and 0.4 tonnes of cardboard were used for packaging of 412 chemo DSC that were
157
continued to be used (no cardboard is used for RSC packaging). During the RSC study year,
158
approximately 60 RSC required repair with 30 kg parts being recycled (80%) or reused (20%) (nil to
159
landfill), and, with recycling credit, an equivalent of 3.7 RSC were manufactured as replacement
160
containers (2,783 RSC total for year). In the RSC study-year, the manufacture, treatment and
PeerJ Preprints | https://doi.org/10.7287/peerj.preprints.26517v2 | CC BY 4.0 Open Access | rec: 23 May 2018, publ: 23 May 2018
161
disposal of 412 chemotherapy DSC were included. The RSC in this study, certified for 500 uses,
162
were reused an average of 12.0 times/year, giving a theoretical “end-of-life” lifespan of 41.7 years.
163
However a “worst-case” actual lifespan scenario was adopted based on the number of reuses of the
164
most frequently used RSC still in service (each individual RSC is barcoded and uses monitored).
165
Information supplied by the manufacturer stated their most frequently used RSC still in service was
166
19 years old and had been used 360 times, thus giving an expected “worst-case” lifespan of 26.4
167
years. Manufacturing GHG for RSC (calculated by dividing total manufacturing GHG by life
168
expectancy) was 1135 kg CO2eq for a lifespan of 41.7 years (1.3% of total RSC life-cycle GHG) and
169
1795 kgCO2eq for a worst-case lifespan of 26.4 years (2.1% of total RSC life-cycle GHG). The
170
shorter, worst-case lifespan was used in this study. Total GHG emissions and GHG differences
171
between DSC and RSC life cycle stages are shown in Figure 2.
172
Adjusting for the 0.3% APD workload increase in the year of RSC use, sharps management GWP
173
using DSC was 248.6 MTCO2eq, and with RSC use, decreased to 86.20 MTCO2eq, a 162.4
174
MTCO2eq reduction in GWP (65.3%, p